INDEX

Explanations

phrases indicating dependency or conditional relationships

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

見えます

-0.68

Slf

-0.68

行きます

-0.67

らう

-0.65

 Holtz

-0.65

 /*#__

-0.65

nolia

-0.65

LIM

-0.64

labili

-0.64

行きたい

-0.64

POSITIVE LOGITS

で

1.21

で

1.18

して

1.06

円で

1.00

 kaynağından

0.97

んで

0.94

으로

0.93

って

0.92

くて

0.90

りで

0.89

Activations Density 0.032%

phrases indicating dependency or conditional relationships

No Comments

No Known Activations